Development and Evaluation of an Automated e‑Counselling System for Emotion and Sentiment Analysis
pp1‑19
Emmanuel Awuni Kolog, Calkin Suero Montero, Markku Tukiainen
Look inside Download PDF (free)
Abstract
Given the challenges associated with the analysis of emotions in text by counsellors, we present an intelligent e‑counselling system for automatic detection of emotions and sentiments in text. The system‑ EmoTect‑ was developed using a supervised support vector machine learning classifier. Therefore, students’ life stories were collected and developed into a corpus for training and evaluating of the classifier. EmoTect allows users to label instances of the training data based on their own perception of emotions, and then gradually learns to classify emotions according to the user’s perceptions. The EmoTect interface provides a visualization of the emotional changes from automatically analysed students’ submissions over a selectable period. In this paper, the EmoTect classifier is evaluated with a gold standard corpus obtained from students but annotated by counsellors. In addition to the classifier evaluation, the EmoTect prototype was evaluated with counsellors in their settings. From the experimental results, the EmoTect classifier for the sentiment classification achieved comparable accuracy to that achieved with a gold standard when presented with unknown data. The contextual evaluation of the system indicates counsellors' satisfaction and sense of enthusiasm for using EmoTect for counseling delivery.
Keywords: Counselling, Design science research, Emotion classification, Evaluation, Sentiment analysis, Support vector machine
Share this item:
Alexander Ekow Asmah, Joshua Ofoeda, Ken Gyapong
Look inside Download PDF (free)
Abstract
The widespread use of bank cheques in daily life makes the development of cheque processing systems of fundamental relevance to banks and other financial institutions. Few studies on nature of Electronic Cheque Clearing System (ECCS) have also shown jurisdictional differences in the application of the technology. Despite its increasing adoption in Africa, no attempt has been made to model the process and highlight the challenges to drive policy changes. This study addresses this research gap by studying the nature of ECCS in Ghana to identify the inherent challenges. To achieve the objective, data was collected through interviews, observations and direct participation. Findings suggest that cheques go through five set of processes before they are cleared electronically; different from the four processes indicated in existing literature. These processes are Pre‑Conversion, Conversion, Transaction, Security and Storage. The pre‑conversion process depends on the banks objective, whilst some banks centralise the process to reduce cost, other banks decentralise the process to enhance service delivery to customers. The remaining processes are the same across all banks in Ghana. The study also found that security breaches at cheque printing houses, poor bank collaboration and manual cheque reviewing process are the main challenges facing ECCs.
Share this item:
Kennedy Okong’o, Michael Kyobe
Look inside Download PDF (free)
Abstract
In the last two decades, Information Communication Technologies (ICTs) have become a strategic tool of management in developing countries. In specific, Electronic Government enhances governance in the public sector; e‑Government being the use of ICTs in public service. Though information systems as a discipline presents some research on the value of ICTs in the private sector, the focus on the public sector is comparatively thin. Informed by the model of public value management, the paper sought to identify the dimensions of the public value of e‑Government. Thus, a framework was adapted and tested on data collected in a survey of 340 public service officers in Kenya. Through structural equation modeling, an e‑Government public value model was generated, and this formed the main contribution of the paper. At a theoretical level, the model demonstrated cognizance of e‑governance multi‑faceted nature, and as such may inform the development of full‑bodied policies to drive efficiency in public service delivery. The model may aid in elucidation of the drivers which inform the use, or fear of use of e‑government infrastructure. In addition, at a methodological level, the paper suggests the place of mixed methods in information systems research. This aids in understanding the unique qualitative and quantitative measures of perception of public value of e‑Government. In this regard, the estimated model shows the magnitude of influence of e‑Government on various dimensions of public values. In practice, these present a suitable reference to guide the formulation and restructuring of e‑Governance policies and strategies in the developing countries. Though the paper presents a positivistic evidence, it is imprecise on whether the evident values enhance or deteriorate public service quality; thus, possible future research is suggested.
Keywords: Developing Countries, e-Government, Information System, Kenya, Public Value Management, Structural Equations Modeling
Share this item:
Look inside Download PDF (free)
Abstract
SMEs are increasingly pushed to adopt cloud‑based solutions given the risks and costs associated with On‑Premise ERP. Cloud ERP systems are considered to offer high scalability and flexibility since they are proposed as a service. Nevertheless, switching toward such systems remains a hard decision since cloud computing is associated to other kinds of risks such as security, data confidentiality, network dependency, vendor lock‑in, amongst others. While several studies have focused on attitudes and perceptions as the main determinants of switching decisions, this research investigated the effects of personal innovativeness in the IT domain in respect of intentions to switch toward cloud ERP systems with a focus on moderating effects. The author adopted a quantitative approach based on a structured questionnaire administered to IS Managers in Saudi SMEs that actually use On‑Premises ERP. Data collected from 83 firms are analyzed by Structural Equation Modelling using PLS (Partial Least Squares) technique to test the research hypotheses. Among the main results, personal innovativeness is found to significantly moderate the effects of the antecedent (information channels exposure) rather than the effects of the determinants (expected switching risks, satisfaction with current ERP). Unlike previous studies, the results of this study show that personal innovativeness is more crucial when managers develop their perceptions rather than later when deciding about switching toward cloud ERP. From a theoretical perspective, such results steer toward a greater focus on antecedents of intentions to adopt ERP systems. From a managerial perspective, the proposed research model could be used as a framework to help managers assess the opportunities to switch toward cloud ERP.
Keywords: Cloud Computing, Enterprise Resources Planning (ERP), Intention to switch, personal innovativeness, IS Managers
Share this item: